MLCN+TCPW Online Talks
Here, we provide an online platform for scientists working on machine learning with an applied focus in the clinical neurosciences. You find announcements for online talks that we organize together with the TCPW. Usually, these talks take place on a monthly basis. If you are interested in receiving updates via email you can sign up here.
Upcoming TCPW Talks
Social Physiology for Precision Psychiatry
Physiology is the study of how living systems act at the molecular, cellular, and organ system levels. This scientific approach has proven its efficiency to address most of the medical disorders from both qualitative and quantitative standpoints. In psychiatry, however, the diagnosis, prognostic, evaluation, and choice of treatment depend as much on the “milieu social” as the “milieu interieur”. This presentation will argue that recent advances in social neuroscience, systems biology, and digital medicine finally provide all the conceptual and methodological tools to develop a “social physiology” for psychiatry. Moreover, it will illustrate how such a multiscale perspective on mental disorders combined with modern computational tools will enable treatments to be tailored to each patient based on their profiles, from genomes to smartphones.
Cognitive control impairments in depression
Depression is associated with cognitive control impairments. The role of these impairments as potential risk factors for (recurrent) depression will be discussed where there is increasing empirical evidence that impaired cognitive control can be a risk factor for recurrent depression. We discuss recent work on cognitive remediation of cognitive control impairments as a way to reduce relapse and recurrence. Despite advances in the area of training it is becoming increasingly clear that our basic understanding of cognitive control impairments in depression requires improvement. We will discuss recent advances to develop testable theories in this area.
Predicting Cognitive Disorders through Machine and Deep Learning
The human cerebral cortex is implicated in a wide range of neurological and psychiatric disorders. However, current methods for comparing cortical organisation, across populations of controls and patients, remain low powered due to the impact of considerable natural variation of cortical shape and organisation, which confounds standard approaches for analysis based on image registration. In this talk I will describe the challenges, provide a brief tutorial to practices for cortical analysis and then describe new methods developed using surface and volumetric deep learning, which may help us start to address these problems.